Handy TensorFlow.js API for Client-Side ML Development

Let’s look into TensorFlow.js API for training data handling, training execution, and inference. TensorFlow.js is awesome because it brings Machine Learning into the hands of Web developers, this provides mutual benefit. Machine Learning field gets more developers and supporters, while Web development becomes more powerful with the support of Machine Learning.

Read more - Handy TensorFlow.js API for Client-Side ML Development.

Time-Series Prediction Beyond Test Data

I was working on the assignment to build a large scale time-series prediction solution. I end up using a combination of approaches in the single solution — Prophet, ARIMA and LSTM Neural Network (running on top of Keras/TensorFlow). With Prophet (Serving Prophet Model with Flask — Predicting Future) and ARIMA it is straightforward to calculate a prediction for future dates, both provide a function to return prediction for a given future horizon. The same is not obvious with LSTM, if you are new — this will require a significant amount of time to research how to forecast true future dates (most of the examples are showing how to predict against test dataset only).

I found one good example though which I was following and it helped me to solve my task — A Quick Example of Time-Series Prediction Using Long Short-Term Memory (LSTM) Networks. In this post, I will show how to predict shampoo sales monthly data, mainly based on the code from the above example.

Read more - Time-Series Prediction Beyond Test Data.

Publishing Keras Model API with TensorFlow Serving

Building a ML model is a crucial task. Running ML model in production is not a less complex and important task. I had a post in the past about serving ML model through Flask REST API — Publishing Machine Learning API with Python Flask. While this approach works, it certainly lacks some important points:

• Model versioning 
• Request batching 
• Multithreading 

TensorFlow comes with a set of tools to help you run ML model in production. One of these tools — TensorFlow Serving. There is an excellent tutorial that describes how to configure and run it — TensorFlow Serving with Docker. I will follow the same steps in my example.

Read more in my Towards Data Science post.

Multiple Node.js Applications on Oracle Always Free Cloud

What if you want to host multiple Oracle JET applications? You can do it easily on Oracle Always Free Cloud. The solution is described in the below diagram:

You should wrap Oracle JET application into Node.js and deploy it to Oracle Compute Instance through Docker container. This is described in my previous post - Running Oracle JET in Oracle Cloud Free Tier.

Make sure to create Docker container with a port different than 80. To host multiple Oracle JET apps, you will need to create multiple containers, each assigned with a unique port. For example, I'm using port 5000:

docker run -p 5000:3000 -d --name appname dockeruser/dockerimage

This will map standard Node port 3000 to port 5000, accessible internally within Oracle Compute Instance. We can direct external traffic from port 80 to port 5000 (or any other port, mapped with Docker container) through Nginx.

Install Nginx:

yum install nginx

Go to Nginx folder:

cd etc/nginx

Edit configuration file:

nano nginx.conf

Add context root configuration for Oracle JET application, to be directed to local port 5000:

location /invoicingdemoui/ {
     proxy_pass http://127.0.0.1:5000/;
}

To allow HTTP call from Nginx to port 5000 (or other port), run this command (more about it on Stackoverflow):

setsebool -P httpd_can_network_connect 1

Reload Nginx:

systemctl reload nginx

Check Nginx status:

systemctl status nginx

That's all. Your Oracle JET app (demo URL) now accessible from the outside:

Getting Your Hands Dirty with TensorFlow 2.0 and Keras API

Diving into technical details of the regression model creation with TensorFlow 2.0 and Keras API. In TensorFlow 2.0, Keras comes out of the box with TensorFlow library. API is simplified and more convenient to use.

Read the complete article here.

SEO for Single Page Oracle (SPA) JET Application

This is a quick tip post related to Oracle JET.

SEO (search engine optimization) is a must for a public website or public-facing app. Search engines do indexing well for static HTML pages. Oracle JET on contrary follows SPA (single page application) approach with dynamic UI. This makes it harder for the search engines to build an index for such an app. The same applies to any other SPA implementations with Angular, React, Vue.js, etc.

If you want to build SEO support for Oracle JET, similar solutions as for Angular, React or Vue.js SPAs apps can be applied.

One of the simpler and reliable solutions - generate static HTML pages for Oracle JET SPA app. You can follow this article, it explains how the same can be done for Angular - SEO for Single Page Applications.

Machine Learning with SQL

Python (and soon JavaScript with TensorFlow.js) is a dominant language for Machine Learning. What about SQL? There is a way to build/run Machine Learning models in SQL. There could be a benefit to run model training close to the database, where data stays. With SQL we can leverage strong data analysis out of the box and run algorithms without fetching data to the outside world (which could be an expensive operation in terms of performance, especially with large datasets). This post is to describe how to do Machine Learning in the database with SQL.

Read more in my Towards Data Science post.